Adaptive Weighting of Popular Reference Content in Audio Matching

1. A system comprising: a memory that has stored thereon computer executable components; and a processor that executes the following computer executable components stored in the memory: an index component that maintains an audio reference index containing a set of audio references, wherein one or more audio references in the set of audio references is associated with a set of descriptors, and wherein one or more descriptors in the set of descriptors contains a set of interest points; a scoring component that maintains a score associated with the one or more audio references in the set of audio references; and a weighting component that: generates a weighted reference index based on the set of audio references, wherein the set of descriptors is associated with the one or more audio references in the set of audio references, wherein the score is associated with the one or more audio references in the set of audio references, wherein respective ones of the one or more audio references in the weighted reference index are up-weighted or down weighted based at least in part on user popularity; and generates the weighted reference index by adjusting a number of one or more descriptors in a set of descriptors associated with the one or more audio references in the weighted reference index based on the score associated with respective one or more audio references, and allocate an amount of space in memory to store the one or more audio references based on the determination of the weighted reference index.

2. The system of claim 1 , wherein the scoring component also determines the score associated with the one or more audio references based at least in part on at least one of a ranked list of popular recordings, a sales chart of popular songs, a sales chart of popular albums, a ranked list of most frequently played tracks on a radio station, or a ranked list of most frequently played tracks on a set of radio stations.

3. The system of claim 1 , further comprising: an input component that receives an audio signal excerpt; and a matching component that matches the audio signal excerpt to at least one of the one or more audio references among the set of audio references based on the set of descriptors associated with the at least one of the one or more audio references, wherein the scoring component is further configured to increase the score associated with the at least one of the one or more audio references based on the match by the matching component.

4. The system of claim 3 , further comprising a batch component that sends a set of audio signal excerpts to the input component.

5. The system of claim 4 , wherein the scoring component also resets the score associated with the one or more audio references in the set of audio references prior to or contemporaneously with the batch component sending the set of audio signal excerpts to the input component.

6. The system of claim 4 , wherein the set of audio signal excerpts is based on a set of audio signal queries.

7. The system of claim 4 , wherein the set of audio signal queries is based on a period of time.

8. The system of claim 1 , further comprising: a descriptor component that generates at least one interest point associated with at least one of the one or more audio references in the weighted index.

9. The system of claim 8 , wherein the weighting component further replaces interest points in the set of interest points in a descriptor among the set of descriptors associated with the one or more audio references in the weighted reference index with the at least one interest point.

10. The system of claim 8 , wherein the descriptor component further generates at least one additional descriptor of the one or more audio references, wherein the at least one additional descriptor contains the at least one interest point.

11. The system of claim 10 , wherein the weighting component also adjusts the number of the one or more descriptors in the set of descriptors associated with the one or more audio references by associating the at least one additional descriptor of the one or more audio references with one or more audio references.

12. The system of claim 1 , wherein popular queries are weighted higher than unpopular queries in a reference database.

13. The system of claim 12 , wherein the popular queries are determined based at least in part on frequency of matched songs.

14. The system of claim 12 , wherein space allocation in the memory for respective queries is based in part on popularity of respective queries.

15. A method, comprising: using a processor to execute computer executable components stored on a computer readable medium to perform acts comprising: maintaining an audio reference index containing a set of audio references, wherein one or more of the audio references in the set of audio references is associated with a set of descriptors, and wherein one or more descriptors in the set of descriptors contains a set of interest points; maintaining a score associated with the one or more audio references in the set of audio references; and generating a weighted reference index based on the set of audio references, wherein the set of descriptors is associated with the one or more audio references in the set of audio references, wherein the score is associated with the one or more audio references in the set of audio references, wherein respective ones of the one or more audio references in the weighted reference index are up-weighted or down-weighted based at least in part on user popularity, and wherein the generating the weighted reference index is also performed by adjusting a number of the one or more descriptors in the set of descriptors associated with the one or more audio references in the weighted reference index based on the score associated with the respective ones of the one or more audio references, and allocate an amount of space in memory to store the one or more audio references based on the determination of the weighted reference index.

16. The method of claim 15 , wherein the maintaining the score includes adjusting the score associated with the one or more audio references based on at least one of a ranked list of popular recordings, a sales chart of popular songs, a sales chart of popular albums, a ranked list of most frequently played tracks on a radio station, or a ranked list of most frequently played tracks on a set of radio stations.

17. The method of claim 15 , further comprising: receiving a set of audio signal excerpts; and matching one or more audio signal excerpts within the set of audio signal excerpts to the one or more audio references among the set of audio references based on the set of descriptors associated with the one or more audio references, wherein the score associated with the one or more audio references is adjusted based on the matching.

18. The method of claim 17 , wherein the score associated with one or more audio references in the set of audio references is reset prior to matching the one or more audio signal excerpts within the set of audio signal excerpts.

19. The method of claim 17 , wherein the set of audio signal excerpts are based on a set of audio signal queries.

20. The method of claim 17 , wherein the set of audio signal queries are based on a period of time.

21. The method of claim 15 , further comprising: generating at least one interest point associated with at least one of the one or more audio references in the weighted reference index.

22. The method of claim 21 , further comprising: replacing interest points in a set of interest points in at least one descriptor among the set of descriptors associated with the one or more audio references in the weighted reference index with the at least one interest point.

23. The method of claim 21 , further comprising: generating at least one additional descriptor of the one or more audio references, wherein the at least one additional descriptor contains the at least one interest point.

24. The method of claim 23 , wherein the adjusting the number of the one or more descriptors in the set of descriptors associated with the one or more audio references by associating the at least one additional descriptor of the one or more audio references with the one or more audio references.

25. The method of claim 15 , further comprising weighting higher in a reference database popular queries over unpopular queries.

26. The method of claim 15 , further comprising determining popular queries based at least in part on frequency of matched songs.

27. The method of claim 15 , further comprising allocating space in the memory for respective queries based in part on popularity of respective queries.

28. A non-transitory computer-readable medium storing computer-executable instructions that, in response to execution, cause a device comprising a processor to perform operations, comprising: maintaining an audio reference index containing a set of audio references, wherein one or more audio references in the set of audio references is associated with a set of descriptors; maintaining a score associated with at least one of the one or more audio references in the set of audio references; and generating a weighted reference index based on the set of audio references, wherein the set of descriptors is associated with the one or more audio references in the set of audio references, wherein the score is associated with the one or more audio references in the set of audio references, and wherein respective ones of the one or more audio references are up-weighted or down-weighted based at least in part on user popularity, and wherein the generating the weighted reference index is also performed by adjusting a number of the one or more descriptors in the set of descriptors associated with the one or more audio references in the weighted reference index based on the score associated with the respective ones of the one or more audio references, and allocate an amount of space in memory to store the one or more audio references based on the determination of the weighted reference index.